Temperature-Dependent Demographic Characteristics and Control Potential of Aphelinus asychis Reared from Sitobion avenae as a Biological Control Agent for Myzus persicae on Chili Peppers

Aphelinus asychis, a polyphagous parasitoid, has been widely used as an efficient biological control agent against the aphid Myzus persicae. Aiming to evaluate the influence of temperature on the biological characteristics and control potential of A. asychis for M. persicae, we compared the life table parameters and control potential of A. asychis, which included the developmental time, longevity, fecundity, intrinsic rate of increase (r), and finite killing rate (θ). The results showed that increasing the temperature significantly decreased the developmental time and longevity of A. asychis. The r at 24 (0.2360 d−1) and 28 °C (0.2441 d−1) were significantly greater than those at 20 (0.1848 d−1) and 32 °C (0.1676 d−1). The θ at 24 (0.4495), 28 (0.5414), and 32 °C (0.4312) were also significantly greater than that at 20 °C (0.3140). The relationship between population fitness (r and θ) and temperature followed a unary quadratic function (R2 > 0.95). The temperatures for the expected maximum intrinsic rate of increase (rmax) and the maximum finite killing rate (θmax) were 25.7 and 27.4 °C, respectively. In conclusion, A. asychis could develop and produce progenies within the temperature range of 20–32 °C, and its control efficiency for M. persicae at 24, 28, and 32 °C was greater than that at 20 °C. The most suitable temperature range for controlling M. persicae with A. asychis in the field might be between 25.7 and 27.4 °C.


Introduction
The chili pepper (Capsicum annuum, Solanaceae) is an important vegetable and condiment planted in greenhouses and open-air fields in China [1]. The green peach aphid (Myzus persicae) is a sucking pest of more than 400 host plant species covering 40 families, including the chili pepper [2,3]. It is also an important vector of more than 100 plant viruses [2]. Its fast development and high fecundity promote the build-up of large populations within a short period, especially in greenhouse agroecosystems [4]. In the past few decades, the intensive use of chemical insecticides such as carbamate [5], pyrethroid [6], cyclodiene [7], neonicotinoid [8], and many others to control M. persicae has resulted in their development of resistance to these chemicals. Furthermore, chemical insecticides also have tremendous negative impacts on beneficial organisms and the environment [9][10][11].   The age-stage-specific fecundity (f xj ) was the number of parasitoid progeny at age x and stage j. The age-stage-specific host feeding rate (c xj ) was the number of aphid nymphs killed by A. asychis female adults at age x and stage j for feeding. The age-stage-specific non-effective parasitism rate (d xj ) was the number of aphid nymphs parasitized by A. asychis at age x and stage j but for which emergence failed. The age-stage-specific aphid killing rate (p xj ) was the number of aphid nymphs fed on by A. asychis at age x and stage j, and the p xj value is the sum of f xj , c xj , and d xj .

Life Table
Overlaps between stages revealed the different developmental rates among the A. asychis individuals. The age-specific survival rate (l x ) of A. asychis decreased gradually with increasing age. In the parent cohort, 43, 40, 40, and 26 parasitoids successfully emerged as adults at 20, 24, 28, and 32 • C, respectively. The emergence rates of the parent cohort were 86%, 80%, 80%, and 52% at 20, 24, 28, and 32 • C, respectively. The female proportion in the 20 • C treatment was significantly greater than that in the 28 • C treatment. The increase in temperature caused a significant decrease in the developmental time of A. asychis. Its adult longevity, likewise, showed the same trend ( Figure 1 and Table 2).    The f xj curve indicates the number of progeny adults produced by the female at age x and stage j, and f x2 indicates that the female adult is of the second life stage. The f x2 , m x , and l x m x of the female adult showed irregular fluctuations in all treatments. Increasing temperatures significantly decreased the reproduction period. The total number of progeny adults showed a similar trend ( Figure 2 and Table 2).

Population Parameters
The values of r and λ of A. asychis at 24 and 28 °C were significantly greater than those at 20 and 32 °C, respectively, while the R0 and T decreased significantly with increasing temperature (Table 3).

Host Feeding
The Aphelinus asychis eggs, larvae, and pupae are in host bodies all the time, and this host is

Population Parameters
The values of r and λ of A. asychis at 24 and 28 • C were significantly greater than those at 20 and 32 • C, respectively, while the R 0 and T decreased significantly with increasing temperature (Table 3). Table 3. Population parameters, host feeding, non-effective parasitism and aphid killing of A. asychis parasitizing M. persicae on chili pepper at four constant temperatures.

Host Feeding
The Aphelinus asychis eggs, larvae, and pupae are in host bodies all the time, and this host is regarded as the parental parasitism. Therefore, the host feeding rate dose not exist before the female adult stage. All the k x of the adult female of A. asychis showed irregular undulation at 20, 24, 28, and 32 • C. The maximum daily k x at 20, 24, 28, and 32 • C were 2.4, 2.1, 3.0, and 1.3 aphids at ages 47, 22, 18, and 12 d, respectively. The maximum daily values of q x at 20, 24, 28, and 32 • C were 2.1, 1.8, 1.2, and 0.7 aphids at ages 25, 17, 10, and 12 d, respectively. Increasing the temperature decreased the C 0 of the aphids that were killed by A. asychis (Table 3 and Figure 3).

Non-Effective Parasitism
Because A. asychis could not parasitize aphids during the pre-adult stage, there was no non-effective parasitism rate before adult emergence. The g x of A. asychis showed irregular fluctuation in all treatments. The maximum daily g x at 20, 24, 28, and 32 • C were 3.6, 1.7, 4.0, and 3.2 aphids at ages 52, 22, 21, and 15 d, respectively. The maximum daily values of h x at 20, 24, 28, and 32 • C were 1.0, 1.4, 0.8, and 0.7 aphids at ages 25, 19, 13, and 10 d, respectively. Increasing the temperature significantly decreased the N 0 of A. asychis (Table 3 and Figure 4).

Aphid Killing Rate
Because immature A. asychis could not parasitize and feed on aphids, its killing rate during the pre-adult stage could not be determined. The daily u x of the adult females of A. asychis showed irregular undulation at 20, 24, 28, and 32 • C. The maximum u x at 20, 24, 28, and 32 • C were 25.4, 13.7, 16.0, and 5.4 aphids, respectively. The maximum daily values of w x in the 20, 24, 28, and 32 • C treatments were 12.6, 13.0, 6.2, and 3.5 aphids, respectively. The Z 0 of aphids by A. asychis were 222.8, 124.0, 38.6, and 14.0 aphids per individual at 20, 24, 28, and 32 • C, respectively. The θ at 20, 24, 28, and 32 • C were 0.3140, 0.4495, 0.5414, and 0.4312, respectively. The Q p value of A. asychis increased significantly with increasing temperature ( Figure 5 and Table 3).

Relationship between Population Fitness and Temperatures
The relationship between population fitness (R0, C0, r, and θ) and temperature is shown in Figure  6. The relationship between population fitness (the net reproductive rate, net feeding rate, intrinsic rate of increase, and finite aphid killing rate) and temperature followed a unary quadratic function as evidenced by the high coefficient of determination (R 2 ), greater than 0.95. The net reproductive rate and net host feeding rate decreased as the temperature increased within the range 20 to 32 °C. The temperature for the expected maximum intrinsic rate of increase (25.7 °C) was lower than that for the maximum finite killing rate (27.4 °C).

Relationship between Population Fitness and Temperatures
The relationship between population fitness (R 0 , C 0 , r, and θ) and temperature is shown in Figure 6. The relationship between population fitness (the net reproductive rate, net feeding rate, intrinsic rate of increase, and finite aphid killing rate) and temperature followed a unary quadratic function as evidenced by the high coefficient of determination (R 2 ), greater than 0.95. The net reproductive Insects 2020, 11, 475 8 of 14 rate and net host feeding rate decreased as the temperature increased within the range 20 to 32 • C. The temperature for the expected maximum intrinsic rate of increase (25.7 • C) was lower than that for the maximum finite killing rate (27.4 • C). The population projection showed that A. asychis increased much faster at 24 and 28 °C ( Figure  7). Because the A. asychis female does not feed on another host before the adult stage, and male adult does not feed on aphid, the trend of the total population size was different from that of the killing potential. The curve of the female population size showed, however, a similar trend to that of the killing potential.

Discussion
Temperature is a vital factor that affects the population fitness of insects, of which the optimal for various insect species may vary [48,49]. In this study, the population fitness of A. asychis was The population projection showed that A. asychis increased much faster at 24 and 28 • C (Figure 7). Because the A. asychis female does not feed on another host before the adult stage, and male adult does not feed on aphid, the trend of the total population size was different from that of the killing potential. The curve of the female population size showed, however, a similar trend to that of the killing potential. The population projection showed that A. asychis increased much faster at 24 and 28 °C ( Figure  7). Because the A. asychis female does not feed on another host before the adult stage, and male adult does not feed on aphid, the trend of the total population size was different from that of the killing potential. The curve of the female population size showed, however, a similar trend to that of the killing potential.

Discussion
Temperature is a vital factor that affects the population fitness of insects, of which the optimal for various insect species may vary [48,49]. In this study, the population fitness of A. asychis was

Discussion
Temperature is a vital factor that affects the population fitness of insects, of which the optimal for various insect species may vary [48,49]. In this study, the population fitness of A. asychis was evaluated at four constant temperatures. Aphelinus asychis could survive and produce progenies at all four temperatures, but higher fitness (r and θ) was observed at moderate temperature (24 and 28 • C). The fitting of the data to a unary quadratic function showed that the temperatures for the expected maximum intrinsic rate of increase (r max ) and the maximum finite killing rate (θ max ) were 25.7 and 27.4 • C, respectively. In addition, the temperature for the r max of M. persicae was between 20 and 25 • C [50]. Thus, we inferred that the best temperature range for controlling M. persicae with A. asychis as a biological agent in chili pepper fields might be 25.7-27.4 • C.
Numerous factors might affect the developmental time of A. asychis. Aphelinid wasps, in general, have a developmental time of 15-30 days [13,[51][52][53]. Specifically, the developmental times of A. asychis at 23.9 and 32.2 • C have been determined to be 16 and 10 days, respectively [54]. We found that increasing the temperature significantly decreased the developmental time of the A. asychis female and male. The developmental duration of A. asychis females and males was significantly affected by host age when it fed on Aphis gossypii, which was 14.5 d and 14.4 d in 1-2 day old A. gossypii-nymphs, 13.5 d and 13.1 d in 4-5 day old nymphs, and 12.3 d and 12.2 d with A. gossypii adults as the hosts at 25 • C, respectively [53]. Additionally, the developmental times of the A. asychis female and male from the egg to the adult stage and parasitization of A. gossypii at 25 • C were 13.9 and 13.2 d, respectively [55]. When it parasitized A. gossypii at 20, 25, and 30 • C, the developmental times were 20.6, 14.2, and 13.0 d respectively [56]. Differences among these parameters may be attributed to the temperature, host species, and host stage.
In previous studies, both host species and stage were reported to affect the proportions of A. asychis female adults [54,57]. When the Schizaphis graminum nymph was used as a host, the older-aged nymphs produced a higher proportion of A. asychis female progenies [58]. The proportions of A. asychis that parasitized 1-2-day-old and 4-5-day-old A. gossypii nymphs and adults were 47.4%, 41.2%, and 47.7%, respectively [53]. Additionally, the proportion of female adults produced by A. asychis parasitizing a combination of second and third instar A. gossypii nymphs was 51.9% [55]. The temperature under which the parasitoids are reared may also affect sex ratio. For instance, the highest portion of females on Diaeretiella rapae was 70% at 7.2 • C, and the lowest was 50% at 29.4 • C [59]. Kang et al. reported that the percentages of female adults were 71.7%, 65.0%, and 78.8% at 20, 25, and 30 • C, respectively [56]. This was similar to our results.
Many biotic and abiotic factors, including the host species, host plant, and temperature, could affect the longevity of A. asychis. For example, the longevity of A. asychis female adults when they parasitized S. graminum, Rhopalosiphum maidis, or Sipha flava was similar with some variation-about 18 days under greenhouse conditions [54], 20 days when they parasitized S. graminum under field conditions [57], 21 days when they parasitized the second and third instar nymphs of A. gossypii at 25 • C [55], and 23 days with second instar nymphs of M. persicae on chili peppers at 25 • C [13]. In this study, we found that increasing the temperature significantly decreased the total longevity of A. asychis, and the adult longevities of the females and males showed similar responses to temperature.
In this study, the number of progeny adults of the parasitoids decreased significantly as the temperature increased. When they parasitized S. graminum, R. maidis, and S. flava at 23.9, 26.7, 29.4, and 32.2 • C, the number of A. asychis progeny was less than 200 [54]. A. asychis females produced 232.3, 44.7, and 21.1 eggs when they parasitized 1-2-day-old A. gossypii nymphs, adults, and 4-5-day-old nymphs, respectively [53]. When A. gossypii was used as a host, A. asychis females produced an average of 342.9 mummified aphids at 25 • C [55], which was more than that at 24 • C in this study. When it parasitized the second instar nymph of M. persicae on chili peppers at 25 • C, each A. asychis female produced more eggs (414.6) than that (238.6 eggs) recorded at all four constant temperatures in this study [13]. The difference between these studies may be due to many factors, including the host species, parasitoid strains, or host plants.
Moreover, the R 0 of A. asychis in this study showed a significant decrease with a temperature increase, which indicated that the R 0 was negatively influenced by temperature. The regression equation for R 0 and temperature supports this phenomenon. In this study, the temperature for the expected maximum intrinsic rate of increase (25.7 • C) was lower than that for the maximum finite killing rate (27.4 • C). This shows that the different population characteristics (i.e., population growth and parasitism rate) may respond differently to environmental factors.
The host feeding behaviors of A. asychis on some host species have been previously studied. It was noted that aphids fed on by A. asychis females were first paralyzed and usually died after feeding [18]. In addition, A. gossypii nymphs and adults were acceptable for host feeding by A. asychis, and the number of younger instar aphids for host feeding was higher than that of older instars [53]. They speculated that older aphids were larger and richer in nutrients than the younger nymphs, so the female parasitoids needed more young nymphs to obtain nutrients for oogenesis. Additionally, the aphid's defense reactions may lead to the preference of aphelinids for younger hosts [60][61][62]. Furthermore, the average number of A. gossypii infesting cucumbers (Cucumis sativus) and killed by A. asychis by non-reproductive host killing was 73.9 [63]. As shown in our study, an increase in the rearing temperature significantly decreased the C 0 , and the regression equation for C 0 and temperature showed a similar decreasing tendency from 20 to 30 • C. However, the g x showed an irregular variation in the adult stage at all four constant temperatures, and it was higher at mid-term than prophase. This phenomenon indicated that the emergence rate for progenies in the next generation was influenced by female adult age, which was also found in our previous study [13]. In addition, the N 0 of aphids killed by A. asychis decreased significantly with increasing temperature in this study, which suggested that N 0 might be influenced by temperature.
In this study, the r of A. asychis in the 24 and 28 • C treatments were significantly greater than those in the 20 and 32 • C treatments, which suggested that increasing the temperature benefited the r, but a further increase in temperature negatively affected the population parameter. The r of A. asychis feeding on A. gossypii nymphs was 0.255 at 25 • C [55], which was much greater than that at 20 to 32 • C in our study. The differences between the two studies might be influenced by the host species and temperature.
It is well known that the ability of a natural enemy to kill a pest in a lifespan partially represents its control efficiency. For the age-stage distribution of a stable population, the θ was used to compare the control potential of the natural enemy [43,64]. Our research results showed that the θ of A. asychis gradually decreased significantly with increasing temperature. Therefore, the killing potential of A. asychis for M. persicae on chili peppers was affected by temperature, but the most suitable temperature was around 24 • C.
Aphelinus asychis has a wide distribution in Asia, Europe, and North and South America [65], and has been used in Russia, China, South Korea, Japan, and America [13,17,22,59,66], but the climates in these regions and countries vary. Kalinkat et al. suggested that climate change may influence the functional responses of parasitoid-host pairs via temperature [67]. Temperature is the primary abiotic factor of climate change, which may affect insect development, reproduction, parasitizing behavior, distribution range, and biological clock [68][69][70][71]. The capacities of insects to adapt to new environmental conditions might be conferred by either plasticity or genetic evolution [37]. The difference in the population fitness of A. asychis at the four constant temperatures might be related to a similar mechanism. However, the exact mechanism of adaptation to these circumstances in A. asychis remains unclear and should be studied in the future. In addition, the control efficiency of A. asychis for M. persicae under greenhouse and field conditions, in which the temperature could be varied greatly, need to be tested in future research.

Conclusions
Aphelinus asychis could develop from egg to adult and reproduce successfully within a temperature range of 20-32 • C. The intrinsic rate of increase (r) of A. asychis at 24 and 28 • C was greater than that at 20 and 32 • C, and the finite aphid killing rates (θ) at 24, 28, and 32 • C were better than that at 20 • C.
The population projection showed that A. asychis increased much faster at 24 and 28 • C. The results of fitting data showed that the temperatures for the expected maximum intrinsic rate of increase (r max ) and the maximum finite killing rate (θ max ) were 25.7 and 27.4 • C, respectively, which suggested that the most suitable range of temperatures for A. asychis for controlling M. persicae in chili pepper fields might be between 25.7 and 27.4 • C.